A new method estimating sediment discharge using the pipe hydrophone pulses obtained with multiple amplification factors was proposed. In the method, the proportions of sediment that does not collide with the pipe due to salutation and sediment below the detection limit of grain size by the hydrophone are considered. With this method, it is possible to estimate not only the sediment discharge but also the grain size distribution partially. To verify the validity, the simulated sediment discharge and particle size distribution are compared with the observed values. As a result, the new estimation method simulates the observed sediment discharge and grain size distribution reasonably. However, it was shown that the estimation accuracy varied depending on the individual sediment transport events and the ratio of sediment above the detection limit grain size αn,us plays important role. In addition, sediment discharge simulated by the new method was compared with the conventional pulse method. The result indicates that the new method was not always more accurate than the conventional pulse method for individual events. However, while the conventional method has a large variation between events, the new method can stably maintain high accuracy, and it is shown that the average value for the entire period is superior to the conventional method. It was also found that the value of α determined from the grain size distribution needs to be set appropriately to improve the accuracy by this method. In the future, it is a potential solution to measure the grain size distribution in a short period of time. Or, to reduce the dependence on α, it is necessary to widen the range of the amplification factor that can be used in the estimation method. For that purpose, the detection rate that changes due to pulse saturation should be quantified.